X-Git-Url: http://ftp.safe.ca/?p=safe%2Fjmp%2Flinux-2.6;a=blobdiff_plain;f=mm%2Fpercpu.c;h=39f7dfd59585a118f4a8c6be480bf3b866a50f25;hp=4c8a419119dae00a6957001cc5587228d64c3b14;hb=4f92e2586b43a2402e116055d4edda704f911b5b;hpb=a56dbddf06b653ef9c04ca3767f260fd31ccebab diff --git a/mm/percpu.c b/mm/percpu.c index 4c8a419..39f7dfd 100644 --- a/mm/percpu.c +++ b/mm/percpu.c @@ -1,5 +1,5 @@ /* - * linux/mm/percpu.c - percpu memory allocator + * mm/percpu.c - percpu memory allocator * * Copyright (C) 2009 SUSE Linux Products GmbH * Copyright (C) 2009 Tejun Heo @@ -7,13 +7,13 @@ * This file is released under the GPLv2. * * This is percpu allocator which can handle both static and dynamic - * areas. Percpu areas are allocated in chunks in vmalloc area. Each - * chunk is consisted of num_possible_cpus() units and the first chunk - * is used for static percpu variables in the kernel image (special - * boot time alloc/init handling necessary as these areas need to be - * brought up before allocation services are running). Unit grows as - * necessary and all units grow or shrink in unison. When a chunk is - * filled up, another chunk is allocated. ie. in vmalloc area + * areas. Percpu areas are allocated in chunks. Each chunk is + * consisted of boot-time determined number of units and the first + * chunk is used for static percpu variables in the kernel image + * (special boot time alloc/init handling necessary as these areas + * need to be brought up before allocation services are running). + * Unit grows as necessary and all units grow or shrink in unison. + * When a chunk is filled up, another chunk is allocated. * * c0 c1 c2 * ------------------- ------------------- ------------ @@ -22,11 +22,13 @@ * * Allocation is done in offset-size areas of single unit space. Ie, * an area of 512 bytes at 6k in c1 occupies 512 bytes at 6k of c1:u0, - * c1:u1, c1:u2 and c1:u3. Percpu access can be done by configuring - * percpu base registers UNIT_SIZE apart. + * c1:u1, c1:u2 and c1:u3. On UMA, units corresponds directly to + * cpus. On NUMA, the mapping can be non-linear and even sparse. + * Percpu access can be done by configuring percpu base registers + * according to cpu to unit mapping and pcpu_unit_size. * - * There are usually many small percpu allocations many of them as - * small as 4 bytes. The allocator organizes chunks into lists + * There are usually many small percpu allocations many of them being + * as small as 4 bytes. The allocator organizes chunks into lists * according to free size and tries to allocate from the fullest one. * Each chunk keeps the maximum contiguous area size hint which is * guaranteed to be eqaul to or larger than the maximum contiguous @@ -38,15 +40,14 @@ * region and negative allocated. Allocation inside a chunk is done * by scanning this map sequentially and serving the first matching * entry. This is mostly copied from the percpu_modalloc() allocator. - * Chunks are also linked into a rb tree to ease address to chunk - * mapping during free. + * Chunks can be determined from the address using the index field + * in the page struct. The index field contains a pointer to the chunk. * * To use this allocator, arch code should do the followings. * - * - define CONFIG_HAVE_DYNAMIC_PER_CPU_AREA - * * - define __addr_to_pcpu_ptr() and __pcpu_ptr_to_addr() to translate - * regular address to percpu pointer and back + * regular address to percpu pointer and back if they need to be + * different from the default * * - use pcpu_setup_first_chunk() during percpu area initialization to * setup the first chunk containing the kernel static percpu area @@ -54,75 +55,142 @@ #include #include +#include #include +#include #include #include #include #include #include -#include #include +#include #include #include #include +#include #include +#include #define PCPU_SLOT_BASE_SHIFT 5 /* 1-31 shares the same slot */ #define PCPU_DFL_MAP_ALLOC 16 /* start a map with 16 ents */ +/* default addr <-> pcpu_ptr mapping, override in asm/percpu.h if necessary */ +#ifndef __addr_to_pcpu_ptr +#define __addr_to_pcpu_ptr(addr) \ + (void __percpu *)((unsigned long)(addr) - \ + (unsigned long)pcpu_base_addr + \ + (unsigned long)__per_cpu_start) +#endif +#ifndef __pcpu_ptr_to_addr +#define __pcpu_ptr_to_addr(ptr) \ + (void __force *)((unsigned long)(ptr) + \ + (unsigned long)pcpu_base_addr - \ + (unsigned long)__per_cpu_start) +#endif + struct pcpu_chunk { struct list_head list; /* linked to pcpu_slot lists */ - struct rb_node rb_node; /* key is chunk->vm->addr */ int free_size; /* free bytes in the chunk */ int contig_hint; /* max contiguous size hint */ - struct vm_struct *vm; /* mapped vmalloc region */ + void *base_addr; /* base address of this chunk */ int map_used; /* # of map entries used */ int map_alloc; /* # of map entries allocated */ int *map; /* allocation map */ + void *data; /* chunk data */ bool immutable; /* no [de]population allowed */ - struct page **page; /* points to page array */ - struct page *page_ar[]; /* #cpus * UNIT_PAGES */ + unsigned long populated[]; /* populated bitmap */ }; static int pcpu_unit_pages __read_mostly; static int pcpu_unit_size __read_mostly; -static int pcpu_chunk_size __read_mostly; +static int pcpu_nr_units __read_mostly; +static int pcpu_atom_size __read_mostly; static int pcpu_nr_slots __read_mostly; static size_t pcpu_chunk_struct_size __read_mostly; +/* cpus with the lowest and highest unit numbers */ +static unsigned int pcpu_first_unit_cpu __read_mostly; +static unsigned int pcpu_last_unit_cpu __read_mostly; + /* the address of the first chunk which starts with the kernel static area */ void *pcpu_base_addr __read_mostly; EXPORT_SYMBOL_GPL(pcpu_base_addr); -/* optional reserved chunk, only accessible for reserved allocations */ +static const int *pcpu_unit_map __read_mostly; /* cpu -> unit */ +const unsigned long *pcpu_unit_offsets __read_mostly; /* cpu -> unit offset */ + +/* group information, used for vm allocation */ +static int pcpu_nr_groups __read_mostly; +static const unsigned long *pcpu_group_offsets __read_mostly; +static const size_t *pcpu_group_sizes __read_mostly; + +/* + * The first chunk which always exists. Note that unlike other + * chunks, this one can be allocated and mapped in several different + * ways and thus often doesn't live in the vmalloc area. + */ +static struct pcpu_chunk *pcpu_first_chunk; + +/* + * Optional reserved chunk. This chunk reserves part of the first + * chunk and serves it for reserved allocations. The amount of + * reserved offset is in pcpu_reserved_chunk_limit. When reserved + * area doesn't exist, the following variables contain NULL and 0 + * respectively. + */ static struct pcpu_chunk *pcpu_reserved_chunk; -/* offset limit of the reserved chunk */ static int pcpu_reserved_chunk_limit; /* - * One mutex to rule them all. - * - * The following mutex is grabbed in the outermost public alloc/free - * interface functions and released only when the operation is - * complete. As such, every function in this file other than the - * outermost functions are called under pcpu_mutex. - * - * It can easily be switched to use spinlock such that only the area - * allocation and page population commit are protected with it doing - * actual [de]allocation without holding any lock. However, given - * what this allocator does, I think it's better to let them run - * sequentially. + * Synchronization rules. + * + * There are two locks - pcpu_alloc_mutex and pcpu_lock. The former + * protects allocation/reclaim paths, chunks, populated bitmap and + * vmalloc mapping. The latter is a spinlock and protects the index + * data structures - chunk slots, chunks and area maps in chunks. + * + * During allocation, pcpu_alloc_mutex is kept locked all the time and + * pcpu_lock is grabbed and released as necessary. All actual memory + * allocations are done using GFP_KERNEL with pcpu_lock released. In + * general, percpu memory can't be allocated with irq off but + * irqsave/restore are still used in alloc path so that it can be used + * from early init path - sched_init() specifically. + * + * Free path accesses and alters only the index data structures, so it + * can be safely called from atomic context. When memory needs to be + * returned to the system, free path schedules reclaim_work which + * grabs both pcpu_alloc_mutex and pcpu_lock, unlinks chunks to be + * reclaimed, release both locks and frees the chunks. Note that it's + * necessary to grab both locks to remove a chunk from circulation as + * allocation path might be referencing the chunk with only + * pcpu_alloc_mutex locked. */ -static DEFINE_MUTEX(pcpu_mutex); +static DEFINE_MUTEX(pcpu_alloc_mutex); /* protects whole alloc and reclaim */ +static DEFINE_SPINLOCK(pcpu_lock); /* protects index data structures */ static struct list_head *pcpu_slot __read_mostly; /* chunk list slots */ -static struct rb_root pcpu_addr_root = RB_ROOT; /* chunks by address */ /* reclaim work to release fully free chunks, scheduled from free path */ static void pcpu_reclaim(struct work_struct *work); static DECLARE_WORK(pcpu_reclaim_work, pcpu_reclaim); +static bool pcpu_addr_in_first_chunk(void *addr) +{ + void *first_start = pcpu_first_chunk->base_addr; + + return addr >= first_start && addr < first_start + pcpu_unit_size; +} + +static bool pcpu_addr_in_reserved_chunk(void *addr) +{ + void *first_start = pcpu_first_chunk->base_addr; + + return addr >= first_start && + addr < first_start + pcpu_reserved_chunk_limit; +} + static int __pcpu_size_to_slot(int size) { int highbit = fls(size); /* size is in bytes */ @@ -144,30 +212,60 @@ static int pcpu_chunk_slot(const struct pcpu_chunk *chunk) return pcpu_size_to_slot(chunk->free_size); } -static int pcpu_page_idx(unsigned int cpu, int page_idx) +/* set the pointer to a chunk in a page struct */ +static void pcpu_set_page_chunk(struct page *page, struct pcpu_chunk *pcpu) +{ + page->index = (unsigned long)pcpu; +} + +/* obtain pointer to a chunk from a page struct */ +static struct pcpu_chunk *pcpu_get_page_chunk(struct page *page) +{ + return (struct pcpu_chunk *)page->index; +} + +static int __maybe_unused pcpu_page_idx(unsigned int cpu, int page_idx) { - return cpu * pcpu_unit_pages + page_idx; + return pcpu_unit_map[cpu] * pcpu_unit_pages + page_idx; } -static struct page **pcpu_chunk_pagep(struct pcpu_chunk *chunk, - unsigned int cpu, int page_idx) +static unsigned long __maybe_unused pcpu_chunk_addr(struct pcpu_chunk *chunk, + unsigned int cpu, int page_idx) { - return &chunk->page[pcpu_page_idx(cpu, page_idx)]; + return (unsigned long)chunk->base_addr + pcpu_unit_offsets[cpu] + + (page_idx << PAGE_SHIFT); } -static unsigned long pcpu_chunk_addr(struct pcpu_chunk *chunk, - unsigned int cpu, int page_idx) +static void __maybe_unused pcpu_next_unpop(struct pcpu_chunk *chunk, + int *rs, int *re, int end) { - return (unsigned long)chunk->vm->addr + - (pcpu_page_idx(cpu, page_idx) << PAGE_SHIFT); + *rs = find_next_zero_bit(chunk->populated, end, *rs); + *re = find_next_bit(chunk->populated, end, *rs + 1); } -static bool pcpu_chunk_page_occupied(struct pcpu_chunk *chunk, - int page_idx) +static void __maybe_unused pcpu_next_pop(struct pcpu_chunk *chunk, + int *rs, int *re, int end) { - return *pcpu_chunk_pagep(chunk, 0, page_idx) != NULL; + *rs = find_next_bit(chunk->populated, end, *rs); + *re = find_next_zero_bit(chunk->populated, end, *rs + 1); } +/* + * (Un)populated page region iterators. Iterate over (un)populated + * page regions betwen @start and @end in @chunk. @rs and @re should + * be integer variables and will be set to start and end page index of + * the current region. + */ +#define pcpu_for_each_unpop_region(chunk, rs, re, start, end) \ + for ((rs) = (start), pcpu_next_unpop((chunk), &(rs), &(re), (end)); \ + (rs) < (re); \ + (rs) = (re) + 1, pcpu_next_unpop((chunk), &(rs), &(re), (end))) + +#define pcpu_for_each_pop_region(chunk, rs, re, start, end) \ + for ((rs) = (start), pcpu_next_pop((chunk), &(rs), &(re), (end)); \ + (rs) < (re); \ + (rs) = (re) + 1, pcpu_next_pop((chunk), &(rs), &(re), (end))) + /** * pcpu_mem_alloc - allocate memory * @size: bytes to allocate @@ -176,6 +274,9 @@ static bool pcpu_chunk_page_occupied(struct pcpu_chunk *chunk, * kzalloc() is used; otherwise, vmalloc() is used. The returned * memory is always zeroed. * + * CONTEXT: + * Does GFP_KERNEL allocation. + * * RETURNS: * Pointer to the allocated area on success, NULL on failure. */ @@ -215,6 +316,9 @@ static void pcpu_mem_free(void *ptr, size_t size) * New slot according to the changed state is determined and @chunk is * moved to the slot. Note that the reserved chunk is never put on * chunk slots. + * + * CONTEXT: + * pcpu_lock. */ static void pcpu_chunk_relocate(struct pcpu_chunk *chunk, int oslot) { @@ -228,130 +332,87 @@ static void pcpu_chunk_relocate(struct pcpu_chunk *chunk, int oslot) } } -static struct rb_node **pcpu_chunk_rb_search(void *addr, - struct rb_node **parentp) -{ - struct rb_node **p = &pcpu_addr_root.rb_node; - struct rb_node *parent = NULL; - struct pcpu_chunk *chunk; - - while (*p) { - parent = *p; - chunk = rb_entry(parent, struct pcpu_chunk, rb_node); - - if (addr < chunk->vm->addr) - p = &(*p)->rb_left; - else if (addr > chunk->vm->addr) - p = &(*p)->rb_right; - else - break; - } - - if (parentp) - *parentp = parent; - return p; -} - /** - * pcpu_chunk_addr_search - search for chunk containing specified address - * @addr: address to search for + * pcpu_need_to_extend - determine whether chunk area map needs to be extended + * @chunk: chunk of interest * - * Look for chunk which might contain @addr. More specifically, it - * searchs for the chunk with the highest start address which isn't - * beyond @addr. + * Determine whether area map of @chunk needs to be extended to + * accomodate a new allocation. + * + * CONTEXT: + * pcpu_lock. * * RETURNS: - * The address of the found chunk. + * New target map allocation length if extension is necessary, 0 + * otherwise. */ -static struct pcpu_chunk *pcpu_chunk_addr_search(void *addr) +static int pcpu_need_to_extend(struct pcpu_chunk *chunk) { - struct rb_node *n, *parent; - struct pcpu_chunk *chunk; - - /* is it in the reserved chunk? */ - if (pcpu_reserved_chunk) { - void *start = pcpu_reserved_chunk->vm->addr; - - if (addr >= start && addr < start + pcpu_reserved_chunk_limit) - return pcpu_reserved_chunk; - } + int new_alloc; - /* nah... search the regular ones */ - n = *pcpu_chunk_rb_search(addr, &parent); - if (!n) { - /* no exactly matching chunk, the parent is the closest */ - n = parent; - BUG_ON(!n); - } - chunk = rb_entry(n, struct pcpu_chunk, rb_node); + if (chunk->map_alloc >= chunk->map_used + 2) + return 0; - if (addr < chunk->vm->addr) { - /* the parent was the next one, look for the previous one */ - n = rb_prev(n); - BUG_ON(!n); - chunk = rb_entry(n, struct pcpu_chunk, rb_node); - } + new_alloc = PCPU_DFL_MAP_ALLOC; + while (new_alloc < chunk->map_used + 2) + new_alloc *= 2; - return chunk; + return new_alloc; } /** - * pcpu_chunk_addr_insert - insert chunk into address rb tree - * @new: chunk to insert + * pcpu_extend_area_map - extend area map of a chunk + * @chunk: chunk of interest + * @new_alloc: new target allocation length of the area map * - * Insert @new into address rb tree. - */ -static void pcpu_chunk_addr_insert(struct pcpu_chunk *new) -{ - struct rb_node **p, *parent; - - p = pcpu_chunk_rb_search(new->vm->addr, &parent); - BUG_ON(*p); - rb_link_node(&new->rb_node, parent, p); - rb_insert_color(&new->rb_node, &pcpu_addr_root); -} - -/** - * pcpu_extend_area_map - extend area map for allocation - * @chunk: target chunk + * Extend area map of @chunk to have @new_alloc entries. * - * Extend area map of @chunk so that it can accomodate an allocation. - * A single allocation can split an area into three areas, so this - * function makes sure that @chunk->map has at least two extra slots. + * CONTEXT: + * Does GFP_KERNEL allocation. Grabs and releases pcpu_lock. * * RETURNS: - * 0 if noop, 1 if successfully extended, -errno on failure. + * 0 on success, -errno on failure. */ -static int pcpu_extend_area_map(struct pcpu_chunk *chunk) +static int pcpu_extend_area_map(struct pcpu_chunk *chunk, int new_alloc) { - int new_alloc; - int *new; - size_t size; + int *old = NULL, *new = NULL; + size_t old_size = 0, new_size = new_alloc * sizeof(new[0]); + unsigned long flags; - /* has enough? */ - if (chunk->map_alloc >= chunk->map_used + 2) - return 0; - - new_alloc = PCPU_DFL_MAP_ALLOC; - while (new_alloc < chunk->map_used + 2) - new_alloc *= 2; - - new = pcpu_mem_alloc(new_alloc * sizeof(new[0])); + new = pcpu_mem_alloc(new_size); if (!new) return -ENOMEM; - size = chunk->map_alloc * sizeof(chunk->map[0]); - memcpy(new, chunk->map, size); + /* acquire pcpu_lock and switch to new area map */ + spin_lock_irqsave(&pcpu_lock, flags); + + if (new_alloc <= chunk->map_alloc) + goto out_unlock; + + old_size = chunk->map_alloc * sizeof(chunk->map[0]); + memcpy(new, chunk->map, old_size); /* * map_alloc < PCPU_DFL_MAP_ALLOC indicates that the chunk is * one of the first chunks and still using static map. */ if (chunk->map_alloc >= PCPU_DFL_MAP_ALLOC) - pcpu_mem_free(chunk->map, size); + old = chunk->map; chunk->map_alloc = new_alloc; chunk->map = new; + new = NULL; + +out_unlock: + spin_unlock_irqrestore(&pcpu_lock, flags); + + /* + * pcpu_mem_free() might end up calling vfree() which uses + * IRQ-unsafe lock and thus can't be called under pcpu_lock. + */ + pcpu_mem_free(old, old_size); + pcpu_mem_free(new, new_size); + return 0; } @@ -371,6 +432,9 @@ static int pcpu_extend_area_map(struct pcpu_chunk *chunk) * is inserted after the target block. * * @chunk->map must have enough free slots to accomodate the split. + * + * CONTEXT: + * pcpu_lock. */ static void pcpu_split_block(struct pcpu_chunk *chunk, int i, int head, int tail) @@ -406,6 +470,9 @@ static void pcpu_split_block(struct pcpu_chunk *chunk, int i, * * @chunk->map must have at least two free slots. * + * CONTEXT: + * pcpu_lock. + * * RETURNS: * Allocated offset in @chunk on success, -1 if no matching area is * found. @@ -495,6 +562,9 @@ static int pcpu_alloc_area(struct pcpu_chunk *chunk, int size, int align) * Free area starting from @freeme to @chunk. Note that this function * only modifies the allocation map. It doesn't depopulate or unmap * the area. + * + * CONTEXT: + * pcpu_lock. */ static void pcpu_free_area(struct pcpu_chunk *chunk, int freeme) { @@ -530,216 +600,92 @@ static void pcpu_free_area(struct pcpu_chunk *chunk, int freeme) pcpu_chunk_relocate(chunk, oslot); } -/** - * pcpu_unmap - unmap pages out of a pcpu_chunk - * @chunk: chunk of interest - * @page_start: page index of the first page to unmap - * @page_end: page index of the last page to unmap + 1 - * @flush: whether to flush cache and tlb or not - * - * For each cpu, unmap pages [@page_start,@page_end) out of @chunk. - * If @flush is true, vcache is flushed before unmapping and tlb - * after. - */ -static void pcpu_unmap(struct pcpu_chunk *chunk, int page_start, int page_end, - bool flush) -{ - unsigned int last = num_possible_cpus() - 1; - unsigned int cpu; - - /* unmap must not be done on immutable chunk */ - WARN_ON(chunk->immutable); - - /* - * Each flushing trial can be very expensive, issue flush on - * the whole region at once rather than doing it for each cpu. - * This could be an overkill but is more scalable. - */ - if (flush) - flush_cache_vunmap(pcpu_chunk_addr(chunk, 0, page_start), - pcpu_chunk_addr(chunk, last, page_end)); - - for_each_possible_cpu(cpu) - unmap_kernel_range_noflush( - pcpu_chunk_addr(chunk, cpu, page_start), - (page_end - page_start) << PAGE_SHIFT); - - /* ditto as flush_cache_vunmap() */ - if (flush) - flush_tlb_kernel_range(pcpu_chunk_addr(chunk, 0, page_start), - pcpu_chunk_addr(chunk, last, page_end)); -} - -/** - * pcpu_depopulate_chunk - depopulate and unmap an area of a pcpu_chunk - * @chunk: chunk to depopulate - * @off: offset to the area to depopulate - * @size: size of the area to depopulate in bytes - * @flush: whether to flush cache and tlb or not - * - * For each cpu, depopulate and unmap pages [@page_start,@page_end) - * from @chunk. If @flush is true, vcache is flushed before unmapping - * and tlb after. - */ -static void pcpu_depopulate_chunk(struct pcpu_chunk *chunk, int off, int size, - bool flush) -{ - int page_start = PFN_DOWN(off); - int page_end = PFN_UP(off + size); - int unmap_start = -1; - int uninitialized_var(unmap_end); - unsigned int cpu; - int i; - - for (i = page_start; i < page_end; i++) { - for_each_possible_cpu(cpu) { - struct page **pagep = pcpu_chunk_pagep(chunk, cpu, i); - - if (!*pagep) - continue; - - __free_page(*pagep); - - /* - * If it's partial depopulation, it might get - * populated or depopulated again. Mark the - * page gone. - */ - *pagep = NULL; - - unmap_start = unmap_start < 0 ? i : unmap_start; - unmap_end = i + 1; - } - } - - if (unmap_start >= 0) - pcpu_unmap(chunk, unmap_start, unmap_end, flush); -} - -/** - * pcpu_map - map pages into a pcpu_chunk - * @chunk: chunk of interest - * @page_start: page index of the first page to map - * @page_end: page index of the last page to map + 1 - * - * For each cpu, map pages [@page_start,@page_end) into @chunk. - * vcache is flushed afterwards. - */ -static int pcpu_map(struct pcpu_chunk *chunk, int page_start, int page_end) +static struct pcpu_chunk *pcpu_alloc_chunk(void) { - unsigned int last = num_possible_cpus() - 1; - unsigned int cpu; - int err; - - /* map must not be done on immutable chunk */ - WARN_ON(chunk->immutable); - - for_each_possible_cpu(cpu) { - err = map_kernel_range_noflush( - pcpu_chunk_addr(chunk, cpu, page_start), - (page_end - page_start) << PAGE_SHIFT, - PAGE_KERNEL, - pcpu_chunk_pagep(chunk, cpu, page_start)); - if (err < 0) - return err; - } - - /* flush at once, please read comments in pcpu_unmap() */ - flush_cache_vmap(pcpu_chunk_addr(chunk, 0, page_start), - pcpu_chunk_addr(chunk, last, page_end)); - return 0; -} - -/** - * pcpu_populate_chunk - populate and map an area of a pcpu_chunk - * @chunk: chunk of interest - * @off: offset to the area to populate - * @size: size of the area to populate in bytes - * - * For each cpu, populate and map pages [@page_start,@page_end) into - * @chunk. The area is cleared on return. - */ -static int pcpu_populate_chunk(struct pcpu_chunk *chunk, int off, int size) -{ - const gfp_t alloc_mask = GFP_KERNEL | __GFP_HIGHMEM | __GFP_COLD; - int page_start = PFN_DOWN(off); - int page_end = PFN_UP(off + size); - int map_start = -1; - int uninitialized_var(map_end); - unsigned int cpu; - int i; - - for (i = page_start; i < page_end; i++) { - if (pcpu_chunk_page_occupied(chunk, i)) { - if (map_start >= 0) { - if (pcpu_map(chunk, map_start, map_end)) - goto err; - map_start = -1; - } - continue; - } - - map_start = map_start < 0 ? i : map_start; - map_end = i + 1; + struct pcpu_chunk *chunk; - for_each_possible_cpu(cpu) { - struct page **pagep = pcpu_chunk_pagep(chunk, cpu, i); + chunk = kzalloc(pcpu_chunk_struct_size, GFP_KERNEL); + if (!chunk) + return NULL; - *pagep = alloc_pages_node(cpu_to_node(cpu), - alloc_mask, 0); - if (!*pagep) - goto err; - } + chunk->map = pcpu_mem_alloc(PCPU_DFL_MAP_ALLOC * sizeof(chunk->map[0])); + if (!chunk->map) { + kfree(chunk); + return NULL; } - if (map_start >= 0 && pcpu_map(chunk, map_start, map_end)) - goto err; + chunk->map_alloc = PCPU_DFL_MAP_ALLOC; + chunk->map[chunk->map_used++] = pcpu_unit_size; - for_each_possible_cpu(cpu) - memset(chunk->vm->addr + cpu * pcpu_unit_size + off, 0, - size); + INIT_LIST_HEAD(&chunk->list); + chunk->free_size = pcpu_unit_size; + chunk->contig_hint = pcpu_unit_size; - return 0; -err: - /* likely under heavy memory pressure, give memory back */ - pcpu_depopulate_chunk(chunk, off, size, true); - return -ENOMEM; + return chunk; } -static void free_pcpu_chunk(struct pcpu_chunk *chunk) +static void pcpu_free_chunk(struct pcpu_chunk *chunk) { if (!chunk) return; - if (chunk->vm) - free_vm_area(chunk->vm); pcpu_mem_free(chunk->map, chunk->map_alloc * sizeof(chunk->map[0])); kfree(chunk); } -static struct pcpu_chunk *alloc_pcpu_chunk(void) -{ - struct pcpu_chunk *chunk; - - chunk = kzalloc(pcpu_chunk_struct_size, GFP_KERNEL); - if (!chunk) - return NULL; - - chunk->map = pcpu_mem_alloc(PCPU_DFL_MAP_ALLOC * sizeof(chunk->map[0])); - chunk->map_alloc = PCPU_DFL_MAP_ALLOC; - chunk->map[chunk->map_used++] = pcpu_unit_size; - chunk->page = chunk->page_ar; +/* + * Chunk management implementation. + * + * To allow different implementations, chunk alloc/free and + * [de]population are implemented in a separate file which is pulled + * into this file and compiled together. The following functions + * should be implemented. + * + * pcpu_populate_chunk - populate the specified range of a chunk + * pcpu_depopulate_chunk - depopulate the specified range of a chunk + * pcpu_create_chunk - create a new chunk + * pcpu_destroy_chunk - destroy a chunk, always preceded by full depop + * pcpu_addr_to_page - translate address to physical address + * pcpu_verify_alloc_info - check alloc_info is acceptable during init + */ +static int pcpu_populate_chunk(struct pcpu_chunk *chunk, int off, int size); +static void pcpu_depopulate_chunk(struct pcpu_chunk *chunk, int off, int size); +static struct pcpu_chunk *pcpu_create_chunk(void); +static void pcpu_destroy_chunk(struct pcpu_chunk *chunk); +static struct page *pcpu_addr_to_page(void *addr); +static int __init pcpu_verify_alloc_info(const struct pcpu_alloc_info *ai); + +#ifdef CONFIG_NEED_PER_CPU_KM +#include "percpu-km.c" +#else +#include "percpu-vm.c" +#endif - chunk->vm = get_vm_area(pcpu_chunk_size, GFP_KERNEL); - if (!chunk->vm) { - free_pcpu_chunk(chunk); - return NULL; +/** + * pcpu_chunk_addr_search - determine chunk containing specified address + * @addr: address for which the chunk needs to be determined. + * + * RETURNS: + * The address of the found chunk. + */ +static struct pcpu_chunk *pcpu_chunk_addr_search(void *addr) +{ + /* is it in the first chunk? */ + if (pcpu_addr_in_first_chunk(addr)) { + /* is it in the reserved area? */ + if (pcpu_addr_in_reserved_chunk(addr)) + return pcpu_reserved_chunk; + return pcpu_first_chunk; } - INIT_LIST_HEAD(&chunk->list); - chunk->free_size = pcpu_unit_size; - chunk->contig_hint = pcpu_unit_size; - - return chunk; + /* + * The address is relative to unit0 which might be unused and + * thus unmapped. Offset the address to the unit space of the + * current processor before looking it up in the vmalloc + * space. Note that any possible cpu id can be used here, so + * there's no need to worry about preemption or cpu hotplug. + */ + addr += pcpu_unit_offsets[raw_smp_processor_id()]; + return pcpu_get_page_chunk(pcpu_addr_to_page(addr)); } /** @@ -748,17 +694,21 @@ static struct pcpu_chunk *alloc_pcpu_chunk(void) * @align: alignment of area (max PAGE_SIZE) * @reserved: allocate from the reserved chunk if available * - * Allocate percpu area of @size bytes aligned at @align. Might - * sleep. Might trigger writeouts. + * Allocate percpu area of @size bytes aligned at @align. + * + * CONTEXT: + * Does GFP_KERNEL allocation. * * RETURNS: * Percpu pointer to the allocated area on success, NULL on failure. */ -static void *pcpu_alloc(size_t size, size_t align, bool reserved) +static void __percpu *pcpu_alloc(size_t size, size_t align, bool reserved) { - void *ptr = NULL; + static int warn_limit = 10; struct pcpu_chunk *chunk; - int slot, off; + const char *err; + int slot, off, new_alloc; + unsigned long flags; if (unlikely(!size || size > PCPU_MIN_UNIT_SIZE || align > PAGE_SIZE)) { WARN(true, "illegal size (%zu) or align (%zu) for " @@ -766,27 +716,58 @@ static void *pcpu_alloc(size_t size, size_t align, bool reserved) return NULL; } - mutex_lock(&pcpu_mutex); + mutex_lock(&pcpu_alloc_mutex); + spin_lock_irqsave(&pcpu_lock, flags); /* serve reserved allocations from the reserved chunk if available */ if (reserved && pcpu_reserved_chunk) { chunk = pcpu_reserved_chunk; - if (size > chunk->contig_hint || - pcpu_extend_area_map(chunk) < 0) - goto out_unlock; + + if (size > chunk->contig_hint) { + err = "alloc from reserved chunk failed"; + goto fail_unlock; + } + + while ((new_alloc = pcpu_need_to_extend(chunk))) { + spin_unlock_irqrestore(&pcpu_lock, flags); + if (pcpu_extend_area_map(chunk, new_alloc) < 0) { + err = "failed to extend area map of reserved chunk"; + goto fail_unlock_mutex; + } + spin_lock_irqsave(&pcpu_lock, flags); + } + off = pcpu_alloc_area(chunk, size, align); if (off >= 0) goto area_found; - goto out_unlock; + + err = "alloc from reserved chunk failed"; + goto fail_unlock; } +restart: /* search through normal chunks */ for (slot = pcpu_size_to_slot(size); slot < pcpu_nr_slots; slot++) { list_for_each_entry(chunk, &pcpu_slot[slot], list) { if (size > chunk->contig_hint) continue; - if (pcpu_extend_area_map(chunk) < 0) - goto out_unlock; + + new_alloc = pcpu_need_to_extend(chunk); + if (new_alloc) { + spin_unlock_irqrestore(&pcpu_lock, flags); + if (pcpu_extend_area_map(chunk, + new_alloc) < 0) { + err = "failed to extend area map"; + goto fail_unlock_mutex; + } + spin_lock_irqsave(&pcpu_lock, flags); + /* + * pcpu_lock has been dropped, need to + * restart cpu_slot list walking. + */ + goto restart; + } + off = pcpu_alloc_area(chunk, size, align); if (off >= 0) goto area_found; @@ -794,27 +775,46 @@ static void *pcpu_alloc(size_t size, size_t align, bool reserved) } /* hmmm... no space left, create a new chunk */ - chunk = alloc_pcpu_chunk(); - if (!chunk) - goto out_unlock; - pcpu_chunk_relocate(chunk, -1); - pcpu_chunk_addr_insert(chunk); + spin_unlock_irqrestore(&pcpu_lock, flags); - off = pcpu_alloc_area(chunk, size, align); - if (off < 0) - goto out_unlock; + chunk = pcpu_create_chunk(); + if (!chunk) { + err = "failed to allocate new chunk"; + goto fail_unlock_mutex; + } + + spin_lock_irqsave(&pcpu_lock, flags); + pcpu_chunk_relocate(chunk, -1); + goto restart; area_found: + spin_unlock_irqrestore(&pcpu_lock, flags); + /* populate, map and clear the area */ if (pcpu_populate_chunk(chunk, off, size)) { + spin_lock_irqsave(&pcpu_lock, flags); pcpu_free_area(chunk, off); - goto out_unlock; + err = "failed to populate"; + goto fail_unlock; } - ptr = __addr_to_pcpu_ptr(chunk->vm->addr + off); -out_unlock: - mutex_unlock(&pcpu_mutex); - return ptr; + mutex_unlock(&pcpu_alloc_mutex); + + /* return address relative to base address */ + return __addr_to_pcpu_ptr(chunk->base_addr + off); + +fail_unlock: + spin_unlock_irqrestore(&pcpu_lock, flags); +fail_unlock_mutex: + mutex_unlock(&pcpu_alloc_mutex); + if (warn_limit) { + pr_warning("PERCPU: allocation failed, size=%zu align=%zu, " + "%s\n", size, align, err); + dump_stack(); + if (!--warn_limit) + pr_info("PERCPU: limit reached, disable warning\n"); + } + return NULL; } /** @@ -825,10 +825,13 @@ out_unlock: * Allocate percpu area of @size bytes aligned at @align. Might * sleep. Might trigger writeouts. * + * CONTEXT: + * Does GFP_KERNEL allocation. + * * RETURNS: * Percpu pointer to the allocated area on success, NULL on failure. */ -void *__alloc_percpu(size_t size, size_t align) +void __percpu *__alloc_percpu(size_t size, size_t align) { return pcpu_alloc(size, align, false); } @@ -843,10 +846,13 @@ EXPORT_SYMBOL_GPL(__alloc_percpu); * percpu area if arch has set it up; otherwise, allocation is served * from the same dynamic area. Might sleep. Might trigger writeouts. * + * CONTEXT: + * Does GFP_KERNEL allocation. + * * RETURNS: * Percpu pointer to the allocated area on success, NULL on failure. */ -void *__alloc_reserved_percpu(size_t size, size_t align) +void __percpu *__alloc_reserved_percpu(size_t size, size_t align) { return pcpu_alloc(size, align, true); } @@ -856,6 +862,9 @@ void *__alloc_reserved_percpu(size_t size, size_t align) * @work: unused * * Reclaim all fully free chunks except for the first one. + * + * CONTEXT: + * workqueue context. */ static void pcpu_reclaim(struct work_struct *work) { @@ -863,7 +872,8 @@ static void pcpu_reclaim(struct work_struct *work) struct list_head *head = &pcpu_slot[pcpu_nr_slots - 1]; struct pcpu_chunk *chunk, *next; - mutex_lock(&pcpu_mutex); + mutex_lock(&pcpu_alloc_mutex); + spin_lock_irq(&pcpu_lock); list_for_each_entry_safe(chunk, next, head, list) { WARN_ON(chunk->immutable); @@ -872,37 +882,44 @@ static void pcpu_reclaim(struct work_struct *work) if (chunk == list_first_entry(head, struct pcpu_chunk, list)) continue; - rb_erase(&chunk->rb_node, &pcpu_addr_root); list_move(&chunk->list, &todo); } - mutex_unlock(&pcpu_mutex); + spin_unlock_irq(&pcpu_lock); list_for_each_entry_safe(chunk, next, &todo, list) { - pcpu_depopulate_chunk(chunk, 0, pcpu_unit_size, false); - free_pcpu_chunk(chunk); + pcpu_depopulate_chunk(chunk, 0, pcpu_unit_size); + pcpu_destroy_chunk(chunk); } + + mutex_unlock(&pcpu_alloc_mutex); } /** * free_percpu - free percpu area * @ptr: pointer to area to free * - * Free percpu area @ptr. Might sleep. + * Free percpu area @ptr. + * + * CONTEXT: + * Can be called from atomic context. */ -void free_percpu(void *ptr) +void free_percpu(void __percpu *ptr) { - void *addr = __pcpu_ptr_to_addr(ptr); + void *addr; struct pcpu_chunk *chunk; + unsigned long flags; int off; if (!ptr) return; - mutex_lock(&pcpu_mutex); + addr = __pcpu_ptr_to_addr(ptr); + + spin_lock_irqsave(&pcpu_lock, flags); chunk = pcpu_chunk_addr_search(addr); - off = addr - chunk->vm->addr; + off = addr - chunk->base_addr; pcpu_free_area(chunk, off); @@ -917,34 +934,353 @@ void free_percpu(void *ptr) } } - mutex_unlock(&pcpu_mutex); + spin_unlock_irqrestore(&pcpu_lock, flags); } EXPORT_SYMBOL_GPL(free_percpu); /** - * pcpu_setup_first_chunk - initialize the first percpu chunk - * @get_page_fn: callback to fetch page pointer - * @static_size: the size of static percpu area in bytes + * is_kernel_percpu_address - test whether address is from static percpu area + * @addr: address to test + * + * Test whether @addr belongs to in-kernel static percpu area. Module + * static percpu areas are not considered. For those, use + * is_module_percpu_address(). + * + * RETURNS: + * %true if @addr is from in-kernel static percpu area, %false otherwise. + */ +bool is_kernel_percpu_address(unsigned long addr) +{ + const size_t static_size = __per_cpu_end - __per_cpu_start; + void __percpu *base = __addr_to_pcpu_ptr(pcpu_base_addr); + unsigned int cpu; + + for_each_possible_cpu(cpu) { + void *start = per_cpu_ptr(base, cpu); + + if ((void *)addr >= start && (void *)addr < start + static_size) + return true; + } + return false; +} + +/** + * per_cpu_ptr_to_phys - convert translated percpu address to physical address + * @addr: the address to be converted to physical address + * + * Given @addr which is dereferenceable address obtained via one of + * percpu access macros, this function translates it into its physical + * address. The caller is responsible for ensuring @addr stays valid + * until this function finishes. + * + * RETURNS: + * The physical address for @addr. + */ +phys_addr_t per_cpu_ptr_to_phys(void *addr) +{ + if (pcpu_addr_in_first_chunk(addr)) { + if ((unsigned long)addr < VMALLOC_START || + (unsigned long)addr >= VMALLOC_END) + return __pa(addr); + else + return page_to_phys(vmalloc_to_page(addr)); + } else + return page_to_phys(pcpu_addr_to_page(addr)); +} + +static inline size_t pcpu_calc_fc_sizes(size_t static_size, + size_t reserved_size, + ssize_t *dyn_sizep) +{ + size_t size_sum; + + size_sum = PFN_ALIGN(static_size + reserved_size + + (*dyn_sizep >= 0 ? *dyn_sizep : 0)); + if (*dyn_sizep != 0) + *dyn_sizep = size_sum - static_size - reserved_size; + + return size_sum; +} + +/** + * pcpu_alloc_alloc_info - allocate percpu allocation info + * @nr_groups: the number of groups + * @nr_units: the number of units + * + * Allocate ai which is large enough for @nr_groups groups containing + * @nr_units units. The returned ai's groups[0].cpu_map points to the + * cpu_map array which is long enough for @nr_units and filled with + * NR_CPUS. It's the caller's responsibility to initialize cpu_map + * pointer of other groups. + * + * RETURNS: + * Pointer to the allocated pcpu_alloc_info on success, NULL on + * failure. + */ +struct pcpu_alloc_info * __init pcpu_alloc_alloc_info(int nr_groups, + int nr_units) +{ + struct pcpu_alloc_info *ai; + size_t base_size, ai_size; + void *ptr; + int unit; + + base_size = ALIGN(sizeof(*ai) + nr_groups * sizeof(ai->groups[0]), + __alignof__(ai->groups[0].cpu_map[0])); + ai_size = base_size + nr_units * sizeof(ai->groups[0].cpu_map[0]); + + ptr = alloc_bootmem_nopanic(PFN_ALIGN(ai_size)); + if (!ptr) + return NULL; + ai = ptr; + ptr += base_size; + + ai->groups[0].cpu_map = ptr; + + for (unit = 0; unit < nr_units; unit++) + ai->groups[0].cpu_map[unit] = NR_CPUS; + + ai->nr_groups = nr_groups; + ai->__ai_size = PFN_ALIGN(ai_size); + + return ai; +} + +/** + * pcpu_free_alloc_info - free percpu allocation info + * @ai: pcpu_alloc_info to free + * + * Free @ai which was allocated by pcpu_alloc_alloc_info(). + */ +void __init pcpu_free_alloc_info(struct pcpu_alloc_info *ai) +{ + free_bootmem(__pa(ai), ai->__ai_size); +} + +/** + * pcpu_build_alloc_info - build alloc_info considering distances between CPUs * @reserved_size: the size of reserved percpu area in bytes - * @unit_size: unit size in bytes, must be multiple of PAGE_SIZE, -1 for auto * @dyn_size: free size for dynamic allocation in bytes, -1 for auto - * @base_addr: mapped address, NULL for auto - * @populate_pte_fn: callback to allocate pagetable, NULL if unnecessary + * @atom_size: allocation atom size + * @cpu_distance_fn: callback to determine distance between cpus, optional + * + * This function determines grouping of units, their mappings to cpus + * and other parameters considering needed percpu size, allocation + * atom size and distances between CPUs. + * + * Groups are always mutliples of atom size and CPUs which are of + * LOCAL_DISTANCE both ways are grouped together and share space for + * units in the same group. The returned configuration is guaranteed + * to have CPUs on different nodes on different groups and >=75% usage + * of allocated virtual address space. + * + * RETURNS: + * On success, pointer to the new allocation_info is returned. On + * failure, ERR_PTR value is returned. + */ +struct pcpu_alloc_info * __init pcpu_build_alloc_info( + size_t reserved_size, ssize_t dyn_size, + size_t atom_size, + pcpu_fc_cpu_distance_fn_t cpu_distance_fn) +{ + static int group_map[NR_CPUS] __initdata; + static int group_cnt[NR_CPUS] __initdata; + const size_t static_size = __per_cpu_end - __per_cpu_start; + int group_cnt_max = 0, nr_groups = 1, nr_units = 0; + size_t size_sum, min_unit_size, alloc_size; + int upa, max_upa, uninitialized_var(best_upa); /* units_per_alloc */ + int last_allocs, group, unit; + unsigned int cpu, tcpu; + struct pcpu_alloc_info *ai; + unsigned int *cpu_map; + + /* this function may be called multiple times */ + memset(group_map, 0, sizeof(group_map)); + memset(group_cnt, 0, sizeof(group_map)); + + /* + * Determine min_unit_size, alloc_size and max_upa such that + * alloc_size is multiple of atom_size and is the smallest + * which can accomodate 4k aligned segments which are equal to + * or larger than min_unit_size. + */ + size_sum = pcpu_calc_fc_sizes(static_size, reserved_size, &dyn_size); + min_unit_size = max_t(size_t, size_sum, PCPU_MIN_UNIT_SIZE); + + alloc_size = roundup(min_unit_size, atom_size); + upa = alloc_size / min_unit_size; + while (alloc_size % upa || ((alloc_size / upa) & ~PAGE_MASK)) + upa--; + max_upa = upa; + + /* group cpus according to their proximity */ + for_each_possible_cpu(cpu) { + group = 0; + next_group: + for_each_possible_cpu(tcpu) { + if (cpu == tcpu) + break; + if (group_map[tcpu] == group && cpu_distance_fn && + (cpu_distance_fn(cpu, tcpu) > LOCAL_DISTANCE || + cpu_distance_fn(tcpu, cpu) > LOCAL_DISTANCE)) { + group++; + nr_groups = max(nr_groups, group + 1); + goto next_group; + } + } + group_map[cpu] = group; + group_cnt[group]++; + group_cnt_max = max(group_cnt_max, group_cnt[group]); + } + + /* + * Expand unit size until address space usage goes over 75% + * and then as much as possible without using more address + * space. + */ + last_allocs = INT_MAX; + for (upa = max_upa; upa; upa--) { + int allocs = 0, wasted = 0; + + if (alloc_size % upa || ((alloc_size / upa) & ~PAGE_MASK)) + continue; + + for (group = 0; group < nr_groups; group++) { + int this_allocs = DIV_ROUND_UP(group_cnt[group], upa); + allocs += this_allocs; + wasted += this_allocs * upa - group_cnt[group]; + } + + /* + * Don't accept if wastage is over 25%. The + * greater-than comparison ensures upa==1 always + * passes the following check. + */ + if (wasted > num_possible_cpus() / 3) + continue; + + /* and then don't consume more memory */ + if (allocs > last_allocs) + break; + last_allocs = allocs; + best_upa = upa; + } + upa = best_upa; + + /* allocate and fill alloc_info */ + for (group = 0; group < nr_groups; group++) + nr_units += roundup(group_cnt[group], upa); + + ai = pcpu_alloc_alloc_info(nr_groups, nr_units); + if (!ai) + return ERR_PTR(-ENOMEM); + cpu_map = ai->groups[0].cpu_map; + + for (group = 0; group < nr_groups; group++) { + ai->groups[group].cpu_map = cpu_map; + cpu_map += roundup(group_cnt[group], upa); + } + + ai->static_size = static_size; + ai->reserved_size = reserved_size; + ai->dyn_size = dyn_size; + ai->unit_size = alloc_size / upa; + ai->atom_size = atom_size; + ai->alloc_size = alloc_size; + + for (group = 0, unit = 0; group_cnt[group]; group++) { + struct pcpu_group_info *gi = &ai->groups[group]; + + /* + * Initialize base_offset as if all groups are located + * back-to-back. The caller should update this to + * reflect actual allocation. + */ + gi->base_offset = unit * ai->unit_size; + + for_each_possible_cpu(cpu) + if (group_map[cpu] == group) + gi->cpu_map[gi->nr_units++] = cpu; + gi->nr_units = roundup(gi->nr_units, upa); + unit += gi->nr_units; + } + BUG_ON(unit != nr_units); + + return ai; +} + +/** + * pcpu_dump_alloc_info - print out information about pcpu_alloc_info + * @lvl: loglevel + * @ai: allocation info to dump + * + * Print out information about @ai using loglevel @lvl. + */ +static void pcpu_dump_alloc_info(const char *lvl, + const struct pcpu_alloc_info *ai) +{ + int group_width = 1, cpu_width = 1, width; + char empty_str[] = "--------"; + int alloc = 0, alloc_end = 0; + int group, v; + int upa, apl; /* units per alloc, allocs per line */ + + v = ai->nr_groups; + while (v /= 10) + group_width++; + + v = num_possible_cpus(); + while (v /= 10) + cpu_width++; + empty_str[min_t(int, cpu_width, sizeof(empty_str) - 1)] = '\0'; + + upa = ai->alloc_size / ai->unit_size; + width = upa * (cpu_width + 1) + group_width + 3; + apl = rounddown_pow_of_two(max(60 / width, 1)); + + printk("%spcpu-alloc: s%zu r%zu d%zu u%zu alloc=%zu*%zu", + lvl, ai->static_size, ai->reserved_size, ai->dyn_size, + ai->unit_size, ai->alloc_size / ai->atom_size, ai->atom_size); + + for (group = 0; group < ai->nr_groups; group++) { + const struct pcpu_group_info *gi = &ai->groups[group]; + int unit = 0, unit_end = 0; + + BUG_ON(gi->nr_units % upa); + for (alloc_end += gi->nr_units / upa; + alloc < alloc_end; alloc++) { + if (!(alloc % apl)) { + printk("\n"); + printk("%spcpu-alloc: ", lvl); + } + printk("[%0*d] ", group_width, group); + + for (unit_end += upa; unit < unit_end; unit++) + if (gi->cpu_map[unit] != NR_CPUS) + printk("%0*d ", cpu_width, + gi->cpu_map[unit]); + else + printk("%s ", empty_str); + } + } + printk("\n"); +} + +/** + * pcpu_setup_first_chunk - initialize the first percpu chunk + * @ai: pcpu_alloc_info describing how to percpu area is shaped + * @base_addr: mapped address * * Initialize the first percpu chunk which contains the kernel static * perpcu area. This function is to be called from arch percpu area - * setup path. The first two parameters are mandatory. The rest are - * optional. - * - * @get_page_fn() should return pointer to percpu page given cpu - * number and page number. It should at least return enough pages to - * cover the static area. The returned pages for static area should - * have been initialized with valid data. If @unit_size is specified, - * it can also return pages after the static area. NULL return - * indicates end of pages for the cpu. Note that @get_page_fn() must - * return the same number of pages for all cpus. - * - * @reserved_size, if non-zero, specifies the amount of bytes to + * setup path. + * + * @ai contains all information necessary to initialize the first + * chunk and prime the dynamic percpu allocator. + * + * @ai->static_size is the size of static percpu area. + * + * @ai->reserved_size, if non-zero, specifies the amount of bytes to * reserve after the static area in the first chunk. This reserves * the first chunk such that it's available only through reserved * percpu allocation. This is primarily used to serve module percpu @@ -952,22 +1288,29 @@ EXPORT_SYMBOL_GPL(free_percpu); * limited offset range for symbol relocations to guarantee module * percpu symbols fall inside the relocatable range. * - * @unit_size, if non-negative, specifies unit size and must be - * aligned to PAGE_SIZE and equal to or larger than @static_size + - * @reserved_size + @dyn_size. + * @ai->dyn_size determines the number of bytes available for dynamic + * allocation in the first chunk. The area between @ai->static_size + + * @ai->reserved_size + @ai->dyn_size and @ai->unit_size is unused. + * + * @ai->unit_size specifies unit size and must be aligned to PAGE_SIZE + * and equal to or larger than @ai->static_size + @ai->reserved_size + + * @ai->dyn_size. + * + * @ai->atom_size is the allocation atom size and used as alignment + * for vm areas. * - * @dyn_size, if non-negative, limits the number of bytes available - * for dynamic allocation in the first chunk. Specifying non-negative - * value make percpu leave alone the area beyond @static_size + - * @reserved_size + @dyn_size. + * @ai->alloc_size is the allocation size and always multiple of + * @ai->atom_size. This is larger than @ai->atom_size if + * @ai->unit_size is larger than @ai->atom_size. * - * Non-null @base_addr means that the caller already allocated virtual - * region for the first chunk and mapped it. percpu must not mess - * with the chunk. Note that @base_addr with 0 @unit_size or non-NULL - * @populate_pte_fn doesn't make any sense. + * @ai->nr_groups and @ai->groups describe virtual memory layout of + * percpu areas. Units which should be colocated are put into the + * same group. Dynamic VM areas will be allocated according to these + * groupings. If @ai->nr_groups is zero, a single group containing + * all units is assumed. * - * @populate_pte_fn is used to populate the pagetable. NULL means the - * caller already populated the pagetable. + * The caller should have mapped the first chunk at @base_addr and + * copied static data to each unit. * * If the first chunk ends up with both reserved and dynamic areas, it * is served by two chunks - one to serve the core static and reserved @@ -977,49 +1320,99 @@ EXPORT_SYMBOL_GPL(free_percpu); * and available for dynamic allocation like any other chunks. * * RETURNS: - * The determined pcpu_unit_size which can be used to initialize - * percpu access. + * 0 on success, -errno on failure. */ -size_t __init pcpu_setup_first_chunk(pcpu_get_page_fn_t get_page_fn, - size_t static_size, size_t reserved_size, - ssize_t unit_size, ssize_t dyn_size, - void *base_addr, - pcpu_populate_pte_fn_t populate_pte_fn) +int __init pcpu_setup_first_chunk(const struct pcpu_alloc_info *ai, + void *base_addr) { - static struct vm_struct first_vm; + static char cpus_buf[4096] __initdata; static int smap[2], dmap[2]; + size_t dyn_size = ai->dyn_size; + size_t size_sum = ai->static_size + ai->reserved_size + dyn_size; struct pcpu_chunk *schunk, *dchunk = NULL; + unsigned long *group_offsets; + size_t *group_sizes; + unsigned long *unit_off; unsigned int cpu; - int nr_pages; - int err, i; + int *unit_map; + int group, unit, i; - /* santiy checks */ + cpumask_scnprintf(cpus_buf, sizeof(cpus_buf), cpu_possible_mask); + +#define PCPU_SETUP_BUG_ON(cond) do { \ + if (unlikely(cond)) { \ + pr_emerg("PERCPU: failed to initialize, %s", #cond); \ + pr_emerg("PERCPU: cpu_possible_mask=%s\n", cpus_buf); \ + pcpu_dump_alloc_info(KERN_EMERG, ai); \ + BUG(); \ + } \ +} while (0) + + /* sanity checks */ BUILD_BUG_ON(ARRAY_SIZE(smap) >= PCPU_DFL_MAP_ALLOC || ARRAY_SIZE(dmap) >= PCPU_DFL_MAP_ALLOC); - BUG_ON(!static_size); - if (unit_size >= 0) { - BUG_ON(unit_size < static_size + reserved_size + - (dyn_size >= 0 ? dyn_size : 0)); - BUG_ON(unit_size & ~PAGE_MASK); - } else { - BUG_ON(dyn_size >= 0); - BUG_ON(base_addr); + PCPU_SETUP_BUG_ON(ai->nr_groups <= 0); + PCPU_SETUP_BUG_ON(!ai->static_size); + PCPU_SETUP_BUG_ON(!base_addr); + PCPU_SETUP_BUG_ON(ai->unit_size < size_sum); + PCPU_SETUP_BUG_ON(ai->unit_size & ~PAGE_MASK); + PCPU_SETUP_BUG_ON(ai->unit_size < PCPU_MIN_UNIT_SIZE); + PCPU_SETUP_BUG_ON(pcpu_verify_alloc_info(ai) < 0); + + /* process group information and build config tables accordingly */ + group_offsets = alloc_bootmem(ai->nr_groups * sizeof(group_offsets[0])); + group_sizes = alloc_bootmem(ai->nr_groups * sizeof(group_sizes[0])); + unit_map = alloc_bootmem(nr_cpu_ids * sizeof(unit_map[0])); + unit_off = alloc_bootmem(nr_cpu_ids * sizeof(unit_off[0])); + + for (cpu = 0; cpu < nr_cpu_ids; cpu++) + unit_map[cpu] = UINT_MAX; + pcpu_first_unit_cpu = NR_CPUS; + + for (group = 0, unit = 0; group < ai->nr_groups; group++, unit += i) { + const struct pcpu_group_info *gi = &ai->groups[group]; + + group_offsets[group] = gi->base_offset; + group_sizes[group] = gi->nr_units * ai->unit_size; + + for (i = 0; i < gi->nr_units; i++) { + cpu = gi->cpu_map[i]; + if (cpu == NR_CPUS) + continue; + + PCPU_SETUP_BUG_ON(cpu > nr_cpu_ids); + PCPU_SETUP_BUG_ON(!cpu_possible(cpu)); + PCPU_SETUP_BUG_ON(unit_map[cpu] != UINT_MAX); + + unit_map[cpu] = unit + i; + unit_off[cpu] = gi->base_offset + i * ai->unit_size; + + if (pcpu_first_unit_cpu == NR_CPUS) + pcpu_first_unit_cpu = cpu; + } } - BUG_ON(base_addr && populate_pte_fn); + pcpu_last_unit_cpu = cpu; + pcpu_nr_units = unit; - if (unit_size >= 0) - pcpu_unit_pages = unit_size >> PAGE_SHIFT; - else - pcpu_unit_pages = max_t(int, PCPU_MIN_UNIT_SIZE >> PAGE_SHIFT, - PFN_UP(static_size + reserved_size)); + for_each_possible_cpu(cpu) + PCPU_SETUP_BUG_ON(unit_map[cpu] == UINT_MAX); - pcpu_unit_size = pcpu_unit_pages << PAGE_SHIFT; - pcpu_chunk_size = num_possible_cpus() * pcpu_unit_size; - pcpu_chunk_struct_size = sizeof(struct pcpu_chunk) - + num_possible_cpus() * pcpu_unit_pages * sizeof(struct page *); + /* we're done parsing the input, undefine BUG macro and dump config */ +#undef PCPU_SETUP_BUG_ON + pcpu_dump_alloc_info(KERN_INFO, ai); + + pcpu_nr_groups = ai->nr_groups; + pcpu_group_offsets = group_offsets; + pcpu_group_sizes = group_sizes; + pcpu_unit_map = unit_map; + pcpu_unit_offsets = unit_off; - if (dyn_size < 0) - dyn_size = pcpu_unit_size - static_size - reserved_size; + /* determine basic parameters */ + pcpu_unit_pages = ai->unit_size >> PAGE_SHIFT; + pcpu_unit_size = pcpu_unit_pages << PAGE_SHIFT; + pcpu_atom_size = ai->atom_size; + pcpu_chunk_struct_size = sizeof(struct pcpu_chunk) + + BITS_TO_LONGS(pcpu_unit_pages) * sizeof(unsigned long); /* * Allocate chunk slots. The additional last slot is for @@ -1039,101 +1432,368 @@ size_t __init pcpu_setup_first_chunk(pcpu_get_page_fn_t get_page_fn, */ schunk = alloc_bootmem(pcpu_chunk_struct_size); INIT_LIST_HEAD(&schunk->list); - schunk->vm = &first_vm; + schunk->base_addr = base_addr; schunk->map = smap; schunk->map_alloc = ARRAY_SIZE(smap); - schunk->page = schunk->page_ar; + schunk->immutable = true; + bitmap_fill(schunk->populated, pcpu_unit_pages); - if (reserved_size) { - schunk->free_size = reserved_size; - pcpu_reserved_chunk = schunk; /* not for dynamic alloc */ + if (ai->reserved_size) { + schunk->free_size = ai->reserved_size; + pcpu_reserved_chunk = schunk; + pcpu_reserved_chunk_limit = ai->static_size + ai->reserved_size; } else { schunk->free_size = dyn_size; dyn_size = 0; /* dynamic area covered */ } schunk->contig_hint = schunk->free_size; - schunk->map[schunk->map_used++] = -static_size; + schunk->map[schunk->map_used++] = -ai->static_size; if (schunk->free_size) schunk->map[schunk->map_used++] = schunk->free_size; - pcpu_reserved_chunk_limit = static_size + schunk->free_size; - /* init dynamic chunk if necessary */ if (dyn_size) { - dchunk = alloc_bootmem(sizeof(struct pcpu_chunk)); + dchunk = alloc_bootmem(pcpu_chunk_struct_size); INIT_LIST_HEAD(&dchunk->list); - dchunk->vm = &first_vm; + dchunk->base_addr = base_addr; dchunk->map = dmap; dchunk->map_alloc = ARRAY_SIZE(dmap); - dchunk->page = schunk->page_ar; /* share page map with schunk */ + dchunk->immutable = true; + bitmap_fill(dchunk->populated, pcpu_unit_pages); dchunk->contig_hint = dchunk->free_size = dyn_size; dchunk->map[dchunk->map_used++] = -pcpu_reserved_chunk_limit; dchunk->map[dchunk->map_used++] = dchunk->free_size; } - /* allocate vm address */ - first_vm.flags = VM_ALLOC; - first_vm.size = pcpu_chunk_size; + /* link the first chunk in */ + pcpu_first_chunk = dchunk ?: schunk; + pcpu_chunk_relocate(pcpu_first_chunk, -1); - if (!base_addr) - vm_area_register_early(&first_vm, PAGE_SIZE); - else { - /* - * Pages already mapped. No need to remap into - * vmalloc area. In this case the first chunks can't - * be mapped or unmapped by percpu and are marked - * immutable. - */ - first_vm.addr = base_addr; - schunk->immutable = true; - if (dchunk) - dchunk->immutable = true; - } + /* we're done */ + pcpu_base_addr = base_addr; + return 0; +} - /* assign pages */ - nr_pages = -1; - for_each_possible_cpu(cpu) { - for (i = 0; i < pcpu_unit_pages; i++) { - struct page *page = get_page_fn(cpu, i); +const char *pcpu_fc_names[PCPU_FC_NR] __initdata = { + [PCPU_FC_AUTO] = "auto", + [PCPU_FC_EMBED] = "embed", + [PCPU_FC_PAGE] = "page", +}; - if (!page) - break; - *pcpu_chunk_pagep(schunk, cpu, i) = page; +enum pcpu_fc pcpu_chosen_fc __initdata = PCPU_FC_AUTO; + +static int __init percpu_alloc_setup(char *str) +{ + if (0) + /* nada */; +#ifdef CONFIG_NEED_PER_CPU_EMBED_FIRST_CHUNK + else if (!strcmp(str, "embed")) + pcpu_chosen_fc = PCPU_FC_EMBED; +#endif +#ifdef CONFIG_NEED_PER_CPU_PAGE_FIRST_CHUNK + else if (!strcmp(str, "page")) + pcpu_chosen_fc = PCPU_FC_PAGE; +#endif + else + pr_warning("PERCPU: unknown allocator %s specified\n", str); + + return 0; +} +early_param("percpu_alloc", percpu_alloc_setup); + +#if defined(CONFIG_NEED_PER_CPU_EMBED_FIRST_CHUNK) || \ + !defined(CONFIG_HAVE_SETUP_PER_CPU_AREA) +/** + * pcpu_embed_first_chunk - embed the first percpu chunk into bootmem + * @reserved_size: the size of reserved percpu area in bytes + * @dyn_size: free size for dynamic allocation in bytes, -1 for auto + * @atom_size: allocation atom size + * @cpu_distance_fn: callback to determine distance between cpus, optional + * @alloc_fn: function to allocate percpu page + * @free_fn: funtion to free percpu page + * + * This is a helper to ease setting up embedded first percpu chunk and + * can be called where pcpu_setup_first_chunk() is expected. + * + * If this function is used to setup the first chunk, it is allocated + * by calling @alloc_fn and used as-is without being mapped into + * vmalloc area. Allocations are always whole multiples of @atom_size + * aligned to @atom_size. + * + * This enables the first chunk to piggy back on the linear physical + * mapping which often uses larger page size. Please note that this + * can result in very sparse cpu->unit mapping on NUMA machines thus + * requiring large vmalloc address space. Don't use this allocator if + * vmalloc space is not orders of magnitude larger than distances + * between node memory addresses (ie. 32bit NUMA machines). + * + * When @dyn_size is positive, dynamic area might be larger than + * specified to fill page alignment. When @dyn_size is auto, + * @dyn_size is just big enough to fill page alignment after static + * and reserved areas. + * + * If the needed size is smaller than the minimum or specified unit + * size, the leftover is returned using @free_fn. + * + * RETURNS: + * 0 on success, -errno on failure. + */ +int __init pcpu_embed_first_chunk(size_t reserved_size, ssize_t dyn_size, + size_t atom_size, + pcpu_fc_cpu_distance_fn_t cpu_distance_fn, + pcpu_fc_alloc_fn_t alloc_fn, + pcpu_fc_free_fn_t free_fn) +{ + void *base = (void *)ULONG_MAX; + void **areas = NULL; + struct pcpu_alloc_info *ai; + size_t size_sum, areas_size, max_distance; + int group, i, rc; + + ai = pcpu_build_alloc_info(reserved_size, dyn_size, atom_size, + cpu_distance_fn); + if (IS_ERR(ai)) + return PTR_ERR(ai); + + size_sum = ai->static_size + ai->reserved_size + ai->dyn_size; + areas_size = PFN_ALIGN(ai->nr_groups * sizeof(void *)); + + areas = alloc_bootmem_nopanic(areas_size); + if (!areas) { + rc = -ENOMEM; + goto out_free; + } + + /* allocate, copy and determine base address */ + for (group = 0; group < ai->nr_groups; group++) { + struct pcpu_group_info *gi = &ai->groups[group]; + unsigned int cpu = NR_CPUS; + void *ptr; + + for (i = 0; i < gi->nr_units && cpu == NR_CPUS; i++) + cpu = gi->cpu_map[i]; + BUG_ON(cpu == NR_CPUS); + + /* allocate space for the whole group */ + ptr = alloc_fn(cpu, gi->nr_units * ai->unit_size, atom_size); + if (!ptr) { + rc = -ENOMEM; + goto out_free_areas; } + areas[group] = ptr; - BUG_ON(i < PFN_UP(static_size)); + base = min(ptr, base); - if (nr_pages < 0) - nr_pages = i; - else - BUG_ON(nr_pages != i); + for (i = 0; i < gi->nr_units; i++, ptr += ai->unit_size) { + if (gi->cpu_map[i] == NR_CPUS) { + /* unused unit, free whole */ + free_fn(ptr, ai->unit_size); + continue; + } + /* copy and return the unused part */ + memcpy(ptr, __per_cpu_load, ai->static_size); + free_fn(ptr + size_sum, ai->unit_size - size_sum); + } } - /* map them */ - if (populate_pte_fn) { - for_each_possible_cpu(cpu) - for (i = 0; i < nr_pages; i++) - populate_pte_fn(pcpu_chunk_addr(schunk, - cpu, i)); - - err = pcpu_map(schunk, 0, nr_pages); - if (err) - panic("failed to setup static percpu area, err=%d\n", - err); + /* base address is now known, determine group base offsets */ + max_distance = 0; + for (group = 0; group < ai->nr_groups; group++) { + ai->groups[group].base_offset = areas[group] - base; + max_distance = max_t(size_t, max_distance, + ai->groups[group].base_offset); + } + max_distance += ai->unit_size; + + /* warn if maximum distance is further than 75% of vmalloc space */ + if (max_distance > (VMALLOC_END - VMALLOC_START) * 3 / 4) { + pr_warning("PERCPU: max_distance=0x%zx too large for vmalloc " + "space 0x%lx\n", + max_distance, VMALLOC_END - VMALLOC_START); +#ifdef CONFIG_NEED_PER_CPU_PAGE_FIRST_CHUNK + /* and fail if we have fallback */ + rc = -EINVAL; + goto out_free; +#endif } - /* link the first chunk in */ - if (!dchunk) { - pcpu_chunk_relocate(schunk, -1); - pcpu_chunk_addr_insert(schunk); - } else { - pcpu_chunk_relocate(dchunk, -1); - pcpu_chunk_addr_insert(dchunk); + pr_info("PERCPU: Embedded %zu pages/cpu @%p s%zu r%zu d%zu u%zu\n", + PFN_DOWN(size_sum), base, ai->static_size, ai->reserved_size, + ai->dyn_size, ai->unit_size); + + rc = pcpu_setup_first_chunk(ai, base); + goto out_free; + +out_free_areas: + for (group = 0; group < ai->nr_groups; group++) + free_fn(areas[group], + ai->groups[group].nr_units * ai->unit_size); +out_free: + pcpu_free_alloc_info(ai); + if (areas) + free_bootmem(__pa(areas), areas_size); + return rc; +} +#endif /* CONFIG_NEED_PER_CPU_EMBED_FIRST_CHUNK || + !CONFIG_HAVE_SETUP_PER_CPU_AREA */ + +#ifdef CONFIG_NEED_PER_CPU_PAGE_FIRST_CHUNK +/** + * pcpu_page_first_chunk - map the first chunk using PAGE_SIZE pages + * @reserved_size: the size of reserved percpu area in bytes + * @alloc_fn: function to allocate percpu page, always called with PAGE_SIZE + * @free_fn: funtion to free percpu page, always called with PAGE_SIZE + * @populate_pte_fn: function to populate pte + * + * This is a helper to ease setting up page-remapped first percpu + * chunk and can be called where pcpu_setup_first_chunk() is expected. + * + * This is the basic allocator. Static percpu area is allocated + * page-by-page into vmalloc area. + * + * RETURNS: + * 0 on success, -errno on failure. + */ +int __init pcpu_page_first_chunk(size_t reserved_size, + pcpu_fc_alloc_fn_t alloc_fn, + pcpu_fc_free_fn_t free_fn, + pcpu_fc_populate_pte_fn_t populate_pte_fn) +{ + static struct vm_struct vm; + struct pcpu_alloc_info *ai; + char psize_str[16]; + int unit_pages; + size_t pages_size; + struct page **pages; + int unit, i, j, rc; + + snprintf(psize_str, sizeof(psize_str), "%luK", PAGE_SIZE >> 10); + + ai = pcpu_build_alloc_info(reserved_size, -1, PAGE_SIZE, NULL); + if (IS_ERR(ai)) + return PTR_ERR(ai); + BUG_ON(ai->nr_groups != 1); + BUG_ON(ai->groups[0].nr_units != num_possible_cpus()); + + unit_pages = ai->unit_size >> PAGE_SHIFT; + + /* unaligned allocations can't be freed, round up to page size */ + pages_size = PFN_ALIGN(unit_pages * num_possible_cpus() * + sizeof(pages[0])); + pages = alloc_bootmem(pages_size); + + /* allocate pages */ + j = 0; + for (unit = 0; unit < num_possible_cpus(); unit++) + for (i = 0; i < unit_pages; i++) { + unsigned int cpu = ai->groups[0].cpu_map[unit]; + void *ptr; + + ptr = alloc_fn(cpu, PAGE_SIZE, PAGE_SIZE); + if (!ptr) { + pr_warning("PERCPU: failed to allocate %s page " + "for cpu%u\n", psize_str, cpu); + goto enomem; + } + pages[j++] = virt_to_page(ptr); + } + + /* allocate vm area, map the pages and copy static data */ + vm.flags = VM_ALLOC; + vm.size = num_possible_cpus() * ai->unit_size; + vm_area_register_early(&vm, PAGE_SIZE); + + for (unit = 0; unit < num_possible_cpus(); unit++) { + unsigned long unit_addr = + (unsigned long)vm.addr + unit * ai->unit_size; + + for (i = 0; i < unit_pages; i++) + populate_pte_fn(unit_addr + (i << PAGE_SHIFT)); + + /* pte already populated, the following shouldn't fail */ + rc = __pcpu_map_pages(unit_addr, &pages[unit * unit_pages], + unit_pages); + if (rc < 0) + panic("failed to map percpu area, err=%d\n", rc); + + /* + * FIXME: Archs with virtual cache should flush local + * cache for the linear mapping here - something + * equivalent to flush_cache_vmap() on the local cpu. + * flush_cache_vmap() can't be used as most supporting + * data structures are not set up yet. + */ + + /* copy static data */ + memcpy((void *)unit_addr, __per_cpu_load, ai->static_size); } - /* we're done */ - pcpu_base_addr = (void *)pcpu_chunk_addr(schunk, 0, 0); - return pcpu_unit_size; + /* we're ready, commit */ + pr_info("PERCPU: %d %s pages/cpu @%p s%zu r%zu d%zu\n", + unit_pages, psize_str, vm.addr, ai->static_size, + ai->reserved_size, ai->dyn_size); + + rc = pcpu_setup_first_chunk(ai, vm.addr); + goto out_free_ar; + +enomem: + while (--j >= 0) + free_fn(page_address(pages[j]), PAGE_SIZE); + rc = -ENOMEM; +out_free_ar: + free_bootmem(__pa(pages), pages_size); + pcpu_free_alloc_info(ai); + return rc; +} +#endif /* CONFIG_NEED_PER_CPU_PAGE_FIRST_CHUNK */ + +/* + * Generic percpu area setup. + * + * The embedding helper is used because its behavior closely resembles + * the original non-dynamic generic percpu area setup. This is + * important because many archs have addressing restrictions and might + * fail if the percpu area is located far away from the previous + * location. As an added bonus, in non-NUMA cases, embedding is + * generally a good idea TLB-wise because percpu area can piggy back + * on the physical linear memory mapping which uses large page + * mappings on applicable archs. + */ +#ifndef CONFIG_HAVE_SETUP_PER_CPU_AREA +unsigned long __per_cpu_offset[NR_CPUS] __read_mostly; +EXPORT_SYMBOL(__per_cpu_offset); + +static void * __init pcpu_dfl_fc_alloc(unsigned int cpu, size_t size, + size_t align) +{ + return __alloc_bootmem_nopanic(size, align, __pa(MAX_DMA_ADDRESS)); +} + +static void __init pcpu_dfl_fc_free(void *ptr, size_t size) +{ + free_bootmem(__pa(ptr), size); +} + +void __init setup_per_cpu_areas(void) +{ + unsigned long delta; + unsigned int cpu; + int rc; + + /* + * Always reserve area for module percpu variables. That's + * what the legacy allocator did. + */ + rc = pcpu_embed_first_chunk(PERCPU_MODULE_RESERVE, + PERCPU_DYNAMIC_RESERVE, PAGE_SIZE, NULL, + pcpu_dfl_fc_alloc, pcpu_dfl_fc_free); + if (rc < 0) + panic("Failed to initialized percpu areas."); + + delta = (unsigned long)pcpu_base_addr - (unsigned long)__per_cpu_start; + for_each_possible_cpu(cpu) + __per_cpu_offset[cpu] = delta + pcpu_unit_offsets[cpu]; } +#endif /* CONFIG_HAVE_SETUP_PER_CPU_AREA */